Among the structural changes present in schizophrenia is a decrease in dendritic spine density of pyramidal cells (PCs) in the prefrontal cortex (PFC). Glutamatergic mechanisms are critical to spine development and maintenance. Extracellular glutamate (Glu) levels are determined by a complex interplay between astrocytes and neurons, with astrocytes selectively expressing several proteins that coordinately regulate extracellular Glu. We have found that PFC dopamine (DA) denervation results in a marked decrease in dendritic spine density. We posit that cortical dopamine depletion results in increased neuronally-derived glutamatergic drive onto the dendritic spines of PFC PCs, which sets into play astrocytic mechanisms to homeostatically dampen the increase in Glu and thereby prevent further spine loss and ultimately an excito- toxic loss of neurons. In order to tes this hypothesis, we propose a series of three interrelated specific aims. We will first determine f PFC DA denervation results in changes in key astrocytic factors that modulate extracellular Glu levels, including GLT1, mGluR3, xCT, and kynurenic acid. Changes in astrocytic proteins will be compared to changes in neuronal proteins (including VGluT1 and 2, GluN1 and 2, GluA1 and 2, and EAAC1) involved in glutamatergic transmission. Protein measurements will be performed at one week after DA denervation, when dendritic spine changes have not yet occurred, and three weeks after denervation, a time when dendritic spine numbers are decreased. In the second aim we will determine if extracellular Glu levels, which are derived from both astrocytes and neurons but which do not represent synaptic Glu, are changed in the DA-denervated PFC. We will also perform physiological experiments to monitor neuronally-derived synaptic Glu activity by recording sEPSCs and evoked EPSCs from layer V PCs using whole cell patch clamping. These measurements will be performed at one and three weeks after DA denervation to determine if changes in synaptic Glu precede changes in extracellular Glu. In the final aim we will pharmacologically modify key astrocytic processes and determine if these changes prevent or reverse dendritic spine loss seen in response to DA denervation. As one example, we will determine if treatment of PFC DA-denervated animals with the antibiotic ceftriaxone (CEF), which increases expression of the astrocytic Glu transporter GLT1 and thereby decreases extracellular Glu levels, modifies the effect of DA denervation on spine number. Chronic treatment with CEF will be started on the day after DA denervation to assess prevention of spine loss. We will also start CEF treatment at three weeks after DA denervation, with drug treatment continuing for three weeks, in order to assess if induction of GLT1 attenuates or reverses spine loss once the dsytrophic changes in PC dendrites have occurred. Similarly, we will examine the effects of an mGluR2/3 receptor agonist and kynurenic acid. PUBLIC HEALTH RELEVANCE: The prefrontal cortex (PFC) of persons with schizophrenia is smaller than the PFC in persons without neurological or psychiatric disease, but there is no overall loss of cells. Instead, the architecture of cortical neurons is distorted and the cells los dendritic spines, small protrusions which receive incoming excitatory signals from other neurons. We propose that changes in astrocytes, a type of cell previously thought to be a support cell in the brain but now known to have much broader functions, compensates for the process that causes dendritic spine loss, and that astrocytic proteins may be novel targets for new drug therapies for schizophrenia.